International Journal of Trend in Scientific
Research and Development (IJTSRD)
International Open Access Journal
ISSN No: 2456 - 6470 | www.ijtsrd.com | Volume - 2 | Issue – 4
A Review
eview on Face Detection Techniques
echniques
Ms. A. N. Hire
Hire, Prof. Dr. M. P. Satone
Department of Electronic And Telecommunication
K. K. Wagh Institute of Engineering Education and Research
Nashik, Maharashtra, India
ABSTRACT
From the last two decades, face recognition is playing
an important and vital role especially in the field of
commercial, banking, social and law enforcement
area. It is an interesting application of pattern
recognition and hence received significant attention.
The complete process of face recognition covers in
three stages, face detection, feature extraction and
recognition. Various techniques
ues are then needed for
these three stages. Also these techniques vary from
various other surrounding factors such as face
orientation, expression, lighting and background. This
paper presents the complete study and review of
various techniques used in facee detection and feature
extraction staged under different conditions.
Keywords: Face Recognition, Face
methods, Feature Extraction techniques
Detection
I. INTRODUCTION
Face recognition is a challenging and interesting
research topic in the field of pattern recognition which
has been found a widely used in many applications
such as verification of credit card, security access
control, and human computer interface. Thus many
face recognition algorithms have been proposed and
survey in this area can be found in [2] [3] [4]. There
are two central issues of an automatic face recognition
system; they are (a) feature selection of representation
of face. (b) Classification of new face image based on
the chosen feature representation. Also in a face
recognitionn environment, the result of feature
selection may be affected by some variations in the
face images, such as lighting, expression and pose.
(A )Why
Why use face recognition?
The traditional authentication methods of persons
identity include passwords, PINs, smart cards, plastic
cards, token, keys and so forth. These could be hard to
remember or retain and passwords can be stolen or
guessed, tokens and keys can be misplaced and
forgotten. However an individuals biological trait
tra
cannot be misplaced, forgotten, stolen or forged.
Biometric-based
based technologies include identification
based on physiological characteristics (such as face,
fingerprints, finger geometry, hand geometry, hand
veins, palm, iris, retina, ear and voice) and behavioral
traits (such as gait, signature and keystroke dynamics)
[1]. Face recognition appears to offer several
advantages over other biometric methods. Face
recognition can be done passively without any explicit
action or participation on the part of the
t user since
face images can be acquired from a distance by a
camera. This is particularly beneficial for security and
surveillance purposes. Furthermore, data acquisition
in general is fraught with problems for other
biometrics: techniques that rely on hands
h
and fingers
can be rendered useless if the epidermis tissue is
damaged in some way (i.e., bruised or cracked). Iris
and retina identification require expensive equipment
and are much too sensitive to any body motion. Voice
recognition is susceptible to
o background noises in
public places and auditory fluctuations on a phone
line or tape recording. Signatures can be modified or
forged. However, facial images can be easily obtained
with a couple of inexpensive fixed cameras. Face
recognition is totally non-intrusive
intrusive and does not carry
any such health risks [5].
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 4 | May-Jun
Jun 2018
Page: 1470
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
(B)Applications of face recognition
Face recognition is basically used for two primary
authenticity modes: Verification: Generally described
as one to one matching system because the system
tries to match the image presented the individual
against a specific image already on file. Identification:
It checks the image presented against all others
already in the database. Identification systems are
described as a 1-to-n matching system, where n is the
total number of images in the database. There are
numerous application areas in which face recognition
can be exploited for these two purposes, a few of
which are outlined below. Security (access control to
buildings, airports/seaports, ATM machines and
border checkpoints [12, 13]; computer/ network
security [14]; email authentication on multimedia
workstations).
1) Surveillance: A large number of CCTVs can be
monitored to look for known criminals, drug
offenders, etc. and authorities can be notified
when one is located.
2) General
identity
verification:
Electoral
registration,
banking, electronic commerce,
identifying newborns, national
3) Ds, passports, drivers licenses, employee IDs.
4) Criminal justice systems: mug-shot/booking
systems,
post-event analysis, forensics
5) Image database investigations: Searching image
databases of licensed drivers, benefit recipients,
missing children, immigrants and police bookings
[5].
6) Smart Card applications: In lieu of maintaining a
database of facial images, the face-print can be
stored in a smart card, bar code or magnetic stripe,
authentication of which is performed by matching
the live image and the stored template [7].
7) Multi-media environments with adaptive human
computer interfaces.
8) Video indexing (labeling faces in video) [10, 11]
II.REVIEW OF LITERATURE
Tremendous success being achieved in the fields of
face detection and face recognition, active computing
has received substantial attention among the
researchers in the domain of computer vision. Signals,
which can be used for act recognition, include facial
expression, paralinguistic features of speech, body
language, physiological signals (e.g. Electromyogram
(EMG), Electrocardiogram (ECG), Electrooculogram
(EOG), Electroencephalography (EEG), Functional
Magnetic Resonance Imaging (fMRI) etc.). A review
of signals and methods for active computing is
reported in [1]
Most of the research on facial expression analysis is
based on detection of basic emotions [2]: anger, fear,
disgust, happiness, sadness, and surprise. A number of
novel methodologies for facial expression recognition
have been proposed over the last decade. Active
expression analysis hugely depends upon the accurate
representation of facial features.
facial points is even more challenging than expression
recognition itself. Therefore, most of the existing
algorithms are based on geometric and appearance
based features. The models based on geometric
features track the shape and size of the face and facial
components such as eyes, lip corners, eyebrows etc.,
and categorize the expressions based on relative
position of these facial components.
Some researchers (e.g., [5], [6], [7], [8]) used shape
models based on a set of characteristic points on the
face to classify the expressions. However, these
methods usually require very accurate and reliable
detection as well as tracking of the facial landmarks
which are difficult to achieve in many practical
situations. Moreover, the distance between facial
landmarks vary from person to person, thereby
making the person independent expression
recognition system less reliable. Facial expressions
involve change in local texture.
(c) 3-stages of face recognition Face recognition
technology is a combination of various other
technologies and their features and characteristics
makes face recognition a better performer depending
upon the application. Face recognition works under
three phases- Detection, Extraction and Recognition.
An explanation of each phase of face recognition is
given in the next sections.
Fig. 1. Three main phases of face recognition
problem
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 4 | May-Jun 2018
Page: 1471
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
III.
FACE DETECTION AND ITS
VARIOUS METHODS
It is a fundamental part of the face recognition system
because it has ability to focus computational resources
on the part of an image containing face. Face
detection involves the separation of image into two
parts; one containing the face and the other containing
the background. It is difficult because although
commonalities exist between faces, they can vary
considerably in terms of age, skin color and facial
expression[6].Hjemal and Low [8] divides the face
detection techniques into two categories named
feature based techniques and image based techniques.
(A) Feature based techniques:
The feature based approaches use the facial features to
their detection process. Hjemal and Low [8] further
divide this technique into three categories: low level
analysis, feature analysis and active shape model.
1) Low level analysis: It deals with the segmentation
of visual features by using the properties of pixels,
gray scale level, and motion information. In [9],
implemented an edge representation method for
detecting the facial features in line drawings by
detecting the changes in pixel properties. In [15],
developed this further to detect human head
outline. The edge based techniques rely upon the
labeled edges which are matched to a face model
for verification. Generally eyebrows, pupils and
lips appear darker than surrounding regions, and
thus extraction algorithms can search for local
minima. In contrast, local maxima can be used to
indicate the bright facial spots such as nose tips
[6]. Detection is then performed using low-level
gray-scale thresh holding.
2) Feature analysis: It uses additional knowledge
about the face and removes the ambiguity
produces by low level analysis. The first involves
sequential feature searching strategies based on
the relative positioning of individual facial
features [6]. Initially prominent facial features are
determined which allows less prominent features
to be hypothesized.
3) Active shape models: These are used to define the
actual physical and higher-level appearance of
features. These models are developed by Tim
Cootes and Chris Taylor in 1995. These models
are released near to a feature, such that they
interact with the local image, deforming to take
the shape of the feature [8]. ASM are models of
the shapes of objects which iteratively deform to
fit to an example of the object in a new image. It
works in following two steps: Look in the image
around each point for a better position for that
point, update the model parameters to best match
to these new found positions.
IV. FEATURE EXTRACTION
VARIOUS TECHNIQUES
AND
ITS
Face recognition is an evolving area, changing and
improv-ing constantly. This section gives the
overview of various approaches and techniques along
with their advantages and disadvantages. Different
approaches of face recognition can be categorized in
three main groups such as holistic approach, featurebased approach, and hybrid approach [2]
(A) Geometry based Technique
In this technique feature are extracted using the size
and the relative position of important components of
images. In this technique under the first method firstly
the direction and edges of important component is
detected and then building feature vectors from these
edges and direction. Canny filter and gradient analysis
usually applied in this direction. Second, methods are
based on the grayscales difference of unimportant
components and important components, by using
feature blocks, set of Haar-like feature block in
Adaboost method [20] to change the grayscales
distribution into the feature. In LBP [21] method,
every face image divides into blocks and each block
has its corresponding central pixel. Then this method
examine its neighbor pixels, based on the grayscales
value of central pixel it changes neighbor to 0 or 1.
After that a histograms is build for every region and
then these histograms are combined to a feature vector
for the face image. Technique proposed by Kanade
[22], also comes under this[28].
Fig. 2. Geometric representation of a person
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 4 | May-Jun 2018
Page: 1472
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
(B) Template Based Technique:
This technique extracts facial feature using
appropriate energy function. Methods have been
proposed by Yuille et al. [23], detecting and
parameter value is used. In the Template based first
an eye template is used to detect the eye from
image. Then a correlation is found out between the
eye templates with various overlapping regions of
the face image. Eye region have a maximum
correlation with the template [28].describing
features of faces using deformable templates. In
deformable templates the feature of interest, an eye
for example, is described by a Parameterized
template. These parameterized templates enable a
priori knowledge about the expected shape of the
features to guide the detection process [23]. An
energy function is defined to links peaks, edges, and
valleys in the image intensity with corresponding
properties of the template. After that the template
matching is done with the image, thereby deforming
itself to find the best fit. For the descriptor purpose
final
(C) Appearance Based Approach:
This approach process the image as two
dimensional patterns. The concept of feature in this
approach is different from simple facial features
such as eyes and mouth. Any extracted
characteristic from the image is referred to a
feature. This method group found best performer in
facial feature extraction because it keep the
important infor-mation of image and reject the
redundant information. Method such as principal
component analysis (PCA) and independent
component analysis are used to extract the feature
vector. The main purpose of PCA is to reduce the
large dimensionality of observed variable to the
smaller intrinsic dimensionality of independent
variable without losing much information [25].
Fig. 3. An example of Template based face
recognition
It has been observed that many natural signals,
including speech, natural images, are better described
as linear combinations of sources with super-Gaussian
distributions. In that case, ICA method better than
PCA method because: I) ICA provides a better
probabilistic model of the data. II) It uniquely
identifies the mixing matrix. III) It finds an
unnecessary orthogonal basic which may reconstruct
the data better than PCA in the presence of noise such
as variations lighting and expressions of face. IV) It is
sensitive to high order statistics in the data, not just
the covariance matrix [26] [28].
(D)
Color Based Method:
With the help of different color models like RGB skin
region is detected [29]. The image obtained after
applying skin color statistics is subjected to
binarization. Firstly it is transformed to gray-scale
image and then to a binary image by applying suitable
threshold. All this is done to eliminate the color and
saturation values and consider only the luminance
part. After this luminance part is transformed to
binary image with some threshold because the
features for face are darker than the background
colors. After thresholding noise is removed by
applying some opening and closing operation. Then
eyes, ears, nose facial features can be extracted from
the binary image by considering the threshold for
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 4 | May-Jun 2018
Page: 1473
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
areas which are darker in the mouth than a given
threshold. After getting the triangle, it is easy to get
the coordinates of the four corner points that form the
potential facial region [27][28].
above described techniques. And their comparison
can be concluded as:
Table 1: comparison between various feature
extraction Techniques
Techniq
ue
Metho
ds
Geometry Gabor
based
wavelet
method
V. CONCLUSION
This paper discussed various face detection and
feature extraction techniques in face recognition. Both
are the integral and important part of face recognition
because face classification is totally dependent on
these two. Template based methods are easy to
implement but not represent global face structure.
While color segmentation based methods used color
model for skin detection with morphology operation
to detect features. So different color model and
illumination variation these factors can affect
performance. Appearance based methods represent
optimal feature points which can represent global face
structure. Geometry based methods such as Gabor
wavelet transform face feature extraction provide
stable and scale invariant features.
No. of
featur
e
Eyes,
mouth
and
nose
Advant
ages
Limitati
on
Small
data
base,
recogniti
on rate
95%
Large no.
Of
features
are used
Recogni
tin
rate100
%,simpl
e
manner
Template
based
Deform
able
templat
e
Eyes,
mouth,
nose
and
eyebro
w
Color
based
Color
based
feature
extracti
on
Eyes
and
mouth
Appearan
ce
based
approach
es
PCA,
ICA,
LDA
Eyes
and
mouth
complexi
ty
descriptio
n
b/w
template
and
images
has long
time
Small
Performa
database nce
is
with
Limited
simple
due
to
manner
diversity
Of
backgrou
nd
Small
-need
no. of
good
features quality
recogniti image
on rate
-large
98%
database
require
After studying all the techniques of feature
extractions, we can now conclude the features,
characterstics, advantages and disadvantages of the
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 4 | May-Jun 2018
Page: 1474
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
REFERENCES
1) A. K. Jain, R. Bolle, and S. Pankanti, ”Biometrics:
Personal Identification in Networked Security,” A.
K. Jain, R. Bolle, and S. Pankanti, Eds.: Kluwer
Academic Publishers, 1999.
2) W. Zhao, R. Chellappa, A. Rosenfeld and P. J.
Phillips, Face Recognition: A Literature Survey,
ACM Computing Surveys, vol. 35, No. 4, 2003,
pp.399 - 458
3) Ashok Samal and Prasana A.Iyengar, Automatic
recognition and analysis of human faces and facial
expressions: A survey, Pattern Recognition, vol.
25, 1992, pp.65-77.
4) R. Chellappa, C.L. Wilson, and S. Sirohey, Human
and machine recogni-tion of faces: A survey,
Proceedings of the IEEE, vol. 83, 1995, pp.705740.
5) Rabia Jafri and Hamid R. Arabnia, A Survey of
Face Recognition Techniques, Journal of
Information Processing Systems, Vol.5, No.2,
June 2009
6) Phil Brimblecombe, Face detection using neural
networks.
7) P. J. Phillips, H. Moon, P. J. Rauss, and S. A.
Rizvi, ”The FERET Evaluation Method-ology for
Face Recognition Algorithms,” IEEE Transactions
on Pattern Analysis and Machine Intelligence,
Vol.22,pp.1090-1104, 2000.
8) Erik Hjelmas, Boon Kee Low, Face Detection: A
Survey,
Computer
Vision
and
Image
Understanding, 83, 236-274 April 2001.
9) Toshiyuki Sakai, M. Nagao, Takeo Kanade,
Computer analysis and classification of
photographs of human face, First USA Japan
Computer Conference, 1972.
10) E. Acosta, L. Torres, A. Albiol, and E. J. Delp,
”An automatic face detection and recognition
system for video indexing applications,” in
Proceedings of the IEEE International Conference
on Acoustics, Speech and Signal Processing,
Vol.4. Orlando, Florida, 2002, pp.3644-3647.
11) J.-H. Lee and W.-Y. Kim, ”Video Summarization
and Retrieval System Using Face Recognition and
MPEG-7 Descriptors,” in Image and Video
Retrieval, Vol.3115, Lecture Notes in Computer
Science: Springer Berlin Heidelberg, 2004, pp.179188.
12) K. Kim, ”Intelligent Immigration Control System
by Using Passport Recognition and Face
Verification,” in International Symposium on
Neu-ral Networks. Chongqing, China, 2005,
pp.147-156.
13) J. N. K. Liu, M. Wang, and B. Feng, ”iBotGuard:
an Internet-based intelligent robot security system
using invariant face recognition against intruder, ”
IEEE Transactions on Systems Man And
Cybernetics Part C-Applications And Reviews,
Vol.35, pp.97-105, 2005.
14) H. Moon, ”Biometrics Person Authentication
Using Projection-Based Face Recognition System
in Verification Scenario, ” in International Conference on Bioinformatics and its Applications.
Hong Kong, China, 2004, pp.207-213.
15) Craw, I., Ellis, H. and Lishman, Automatic
extraction of face fea-ture.Pattern Recog. Lett.
183- 187 1987.
16) Rajkiran Kaur, Rachna Rajput, Face recognition
and its various tech-niques: a review, International
Journal of Science, Engineering and Technology
Research (IJSETR) Volume 2, Issue 3, March
2013.
17) L. Wiskott and C. von der Malsburg, Recognizing
faces by dynamic link matching, Neuroimage, vol.
4, pp. 514- 518, 1996
18) H.D. Ellis, Introduction to aspects of face
processing: Ten questions in need of answers, in
Aspects of Face Processing, H. D. Ellis, M.
Jeeves, F. Newcombe, and A. Young Eds.
Dordrecht: Nijhoff, 1986, pp.3-13.
19) Sanjeev Dhawan, Himanshu, A review of Face
Recogntion, IJREAS Volume 2, Issue 2 (February
2012) ISSN: 2249-3905
20) M. Jones and P. Viola, Face Recognition Using
Boosted Local Features, IEEE International
Conference on Computer Vision,2003.
21) Shu Liao, Wei Fan, Albert C. S. Chung and DitYan Yeung, Facial Expression Recognition Using
Advanced Local Binary Patterns, Tsallis Entropies
Erik Hjelmas, Boon Kee Low, Face Detection: A
Survey,
Computer
Vision
and
Image
Understanding, 83, 236-274 April 2001.
22) M. Nixon, Eye spacing measurement for facial
recognition,. Proceedings of the Society of PhotoOptical Instrument Engineers, SPIE, 575(37):279
285, August 1985.
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 4 | May-Jun 2018
Page: 1475
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
23) Yuille, A. L., Cohen, D. S., and Hallinan, P. W.,
”Feature extraction from faces using deformable
templates”, Proc. of CVPR, (1989).
24) N. Bhoi, M. N. Mohanty, Template Matching
based Eye Detection in Facial Image, International
Journal of Computer Applications (0975 8887)
Volume 12 No.5, December 2010
25) M. Turk and A. Pentland. Eigenfaces for
recognition Journal of Cog-nitive Neuroscience,
3(1):7186, 1991
26) J. Lim, Y. Kim J. Paik Comparative Analysis of
WaveletBased
Scale-Invariant
Feature
Extraction Using Different Wavelet Bases
International Journal of Signal Processing, Image
Processing and Pattern Recognition Vol. 2, No. 4,
December, 2009.
27) S. K. Singh, D. S. Chauhan, M. Vatsa, R. Singh,
A Robust Skin Color Based Face Detection
Algorithm, Tamkang Journal of Science and
Engineering, Vol. 6, No. 4, pp. 227-234 (2003).
28) Sanjeev Dhawan, himanshu Dogra, Feature
Extraction Techniques for Face Recognition,
International Journal of Engineering, Business and
Enterprise Applications (IJEBEA), 2012.
29) V.Vezhnevets, V.Sazonov ,A. Andreeva , A
Survey on Pixel- Based Skin Color Detection
Techniques, Graphics and Media Laboratory,
Moscow State University, Moscow, Russia.
30) Vladimir Vezhnevets, Vassili Sazonov, Alla
Andreeva., A Survey on Pixel-Based Skin Color
DetectionTechniques.
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 4 | May-Jun 2018
Page: 1476